A growing consensus suggests that epigenetics, i.e., mitotically heritable information that controls gene expression without a change in the DNA sequence, plays an important role in endocrine adjustment associated with the genesis of obesity and obesity-related adult-onset chronic diseases. Because imprinted genes depend on epigenetic mechanisms such as DNA methylation for their monoallelic expression, variation in methyl-group donor nutrients makes them vulnerable to epigenetic deregulation, especially since only one allele is functional. Deregulation of the paternally expressed IGF2 has been associated with increased risk of obesity, overgrowth disorders and obesity-related cancers in animal and epidemiologic studies. DLK1 exhibits many of the same structural and regulatory characteristics of the coordinately regulated IGF2 imprinted domain. Despite the seeming importance of imprinted genes in obesity and adult-onset chronic disease, the population frequency of, and risk factors associated with deregulation (aberrant methylation and loss of imprinting) of these paternally expressed genes is unknown. Exposure to diets with a high methyl group donor content has been associated with aberrant methylation and cancer. Our central hypothesis is that in utero and early ex utero exposure to `methyl-rich'diets increases the risk of aberrant methylation at imprint control centers of imprinted domains, increasing the risk of loss of imprinting and increased circulating concentrations of growth factors they encode, and subsequently, rapid weight gain in the offspring. Our specific aims are: (1) To determine whether maternal intake of B vitamins-- folate, vitamins B12 and B6--are associated with aberrant methylation and loss of imprinting of insulin-like growth factor-2 (IGF2) and Delta, drosophila homolog-like-1 (DLK1) in the offspring;(2) To determine whether aberrant methylation and loss of imprinting of IGF2 and DLK1 is associated with serum concentrations of IGF2 and DLK1 in infants, and (3) To determine whether early childhood exposure to these B vitamins is associated with aberrant methylation and subsequent weight gain during the first three years of life. To address these aims, we will recruit and estimate micronutrient intake in 1,500 women in the first and second/third trimester of pregnancy. At delivery, we will collect cord blood and buccal cell specimens from which we will evaluate the methylation profile and loss of imprinting in IGF2 and DLK1 in relation to maternal nutrition. At ages one, two and three, we will collect buccal cells, infant data on usual diet, weight and height measurements, to determine whether infant nutrition is associated with an aberrant methylation profile and subsequent infant weight gain. Loss of imprinting is mitotically heritable, but, in the absence of mutation in methylation-regulating genes, can be reversed with normal imprinting restored. Thus, identification of risk factors associated with deregulation of imprint control domains may hold therapeutic and focused public health intervention prospects for childhood obesity and adult onset, obesity-related chronic diseases. PUBLIC HEALTH RELEVANCE: This longitudinal epidemiology project will study how prenatal and early childhood nutrition affect the function of genes believed to play role in the development of obesity and obesity-related chronic disease. Because nutrition-related patterns of gene deregulation are potentially reversible, identifying nutritional and other risk factors associated with abnormal gene functioning may lead to therapeutic and public health interventions for childhood obesity and for adult onset, obesity-related chronic diseases.